Advances in deep learning algorithms overshadow their security risk software implementations. This paper discloses a set of vulnerabilities popular frameworks including Caffe, TensorFlow, and Torch. Contrary to the small code size models, these are complex, they heavily depend on numerous open source packages. considers risks caused by studying impact common applications such as voice recognition image classification. By exploiting framework implementations, attackers can launch...

We consider an inverse source problem for the Helmholtz equation. This is concerned with reconstruction of unknown from multi-frequency data obtained radiated fields. Based on a Fourier expansion source, numerical method proposed to solve problem. Stability analyzed and experiments are presented show effectiveness our method.

We consider an inverse source problem of determining a term in the Helmholtz equation from multi-frequency far-field measurements. Based on Fourier series expansion, we develop novel non-iterative reconstruction method for solving problem. A promising feature this is that it utilizes data only few observation directions each frequency. Theoretical uniqueness and stability analysis are provided. Numerical experiments conducted to illustrate effectiveness efficiency proposed both two three dimensions.

This paper is devoted to the uniqueness in inverse acoustic scattering problems for Helmholtz equation with phaseless far-field data. Some novel techniques are developed overcome difficulty of translation invariance induced by a single incident plane wave. In this paper, based on adding reference ball as an extra artificial impenetrable obstacle (resp. penetrable homogeneous medium) system and then using superpositions fixed wave some point sources waves, we rigorously prove that location...

In this paper, we consider the inverse problem of determining location and shape a sound-soft obstacle from modulus far-field data for single incident plane wave. By adding reference ball artificially to scattering system, propose system nonlinear integral equations based iterative scheme reconstruct both obstacle. The technique causes few extra computational costs, but breaks translation invariance brings information about Several validating numerical examples are provided illustrate...

This work is concerned with the inverse source problem of locating multiple multipolar sources from boundary measurements for Helmholtz equation. We develop simple and effective sampling schemes location acquisition a single wavenumber. Our algorithms are based on some novel indicator functions whose indicating behaviors could be used to locate sources. The inversion totally "direct" in sense that only integral calculations involved evaluating functions. Rigorous mathematical justifications...

This paper is concerned with the inverse source problem of reconstructing an unknown acoustic excitation from phaseless measurements radiated fields away at multiple frequencies. It well known that non-uniqueness issue a major challenge associated such problem. We develop novel strategy to overcome this challenging by recovering via adding some reference point sources as extra artificial system. technique requires only few data, and brings in simple phase retrieval formula. The stability...

This paper concerns the simultaneous reconstruction of a sound-soft cavity and its excitation sources from total-field data. Using single-layer potential representations on two measurement curves, this co-inversion problem can be decoupled into inverse problems: an scattering source problem. novel decoupling technique is fast easy to implement since it based linear system integral equations. Then uncoupled subproblems are respectively solved by modified optimization sampling method. We also...

In this paper, we establish the unique determination results for several inverse acoustic scattering problems using modulus of near-field data. By utilizing superpositions point sources as incident waves, rigorously prove that phaseless near-fields collected on an admissible surface can uniquely determine location and shape obstacle well its boundary condition refractive index a medium inclusion, respectively. We also uniqueness in determining locally rough from data due to sources. These...

This paper is concerned with the uniqueness in inverse acoustic scattering problems modulus of far-field patterns co-produced by obstacle (resp. medium) and point sources. Based on superposition sources as incident waves, we overcome difficulty translation invariance induced a single plane wave, rigorously prove that location shape well its boundary condition or refractive index can be uniquely determined patterns. work different from our previous phaseless [2018 Inverse Problems 34,...

In the context of Industry 4.0, most popular way to identify and track objects is add tags, currently companies still use cheap quick response (QR) which can be positioned by computer vision (CV) technology. CV, instance segmentation (IS) detect position tags while also segmenting each instance. Currently, mask region-based convolutional neural network (Mask R-CNN) method used realize IS, but completeness cannot guaranteed. Furthermore, due rich texture QR low-quality images lower...

We develop a finite element method with rectangular perfectly matched layers (PMLs) for the wave scattering from two-dimensional cavities.The unbounded computational domain is truncated to bounded one by using of layer at open aperture.The PML parameters such as thickness and fictitious medium property are determined through sharp posteriori error estimates.Numerical experiments carried out illustrate competitive behavior proposed method.

Abstract This work is concerned with an inverse elastic scattering problem of identifying the unknown rigid obstacle embedded in open space filled a homogeneous and isotropic medium. A Newton-type iteration method relying on boundary condition designed to identify curve obstacle. Based Helmholtz decomposition Fourier–Bessel expansion, we explicitly derive approximate scattered field its derivative each iterative curve. Rigorous mathematical justifications for proposed are provided. Numerical...

<p style='text-indent:20px;'>This paper is concerned with the inverse acoustic scattering problems by an obstacle or a cavity sound-soft sound-hard boundary. A direct imaging method relying on boundary conditions proposed for reconstructing shape of cavity. First, scattered fields are approximated Fourier-Bessel functions measurements closed curve. Then, indicator established superposition total their derivatives to incident point sources. We prove that vanish only Numerical examples...

The inverse problem considered in this article is to determine the shape of a two-dimensional time-harmonic acoustic scatterer with Dirichlet boundary conditions from knowledge some far field patterns. Based on optimization method due Kirsch and Kress for scattering problem, we propose new scheme by reformulating cost functional via technique piecewise integration respect incident directions. Convergence analysis given. Numerical experiments show that our accelerates computations without...